Rock Mech Rock Eng (2016) 49:2055–2064

DOI 10.1007/s00603-015-0905-9

ORIGINAL PAPER

Experimental Study of Estimating the Subgrade Reaction

Modulus on Jointed Rock Foundations
Jaehwan Lee1 • Sangseom Jeong1

Received: 10 August 2015 / Accepted: 22 December 2015 / Published online: 23 January 2016

p y
o
Ó Springer-Verlag Wien 2016

Abstract The subgrade reaction modulus for rock foun- dulus of subgrade reaction is assumed to be
words, the modulus

c
dations under axial loading is investigated by model foot- the elastic constant off the springs.
ing tests. This study focuses on quantifying a new subgrade Overr 90 % of the shallow foundations constructed
construct in
reaction modulus by considering rock discontinuities. A uth Koreaa are constructed on weathered rocks.
South ro The

r
series of model-scale footing tests are performed to weathered rocks, which occupy two-thirds of the total land
investigate the effects of the unconfined compressive ve penins
area of the Korean peninsula, general the result of the
are generally
strength, discontinuity spacing and inclination of thee rock granite-gne
physical weathering of granite-gneiss of varying thick-

o
served thatt
joint. Based on the experimental results, it is observed nesses up to 40 m. It is clear thatt most rocks cannot be
the subgrade reaction modulus of the rock with disconti- a
accurately is
represented as isotropic linear elastic materials
nuities decreases by up to approximately ely 60 % of intact bec
because presen
of the presence of joints and discontinuities.

h
rock. In addition, it is found that thee modulus of subgrade Disco
Discontinuities ro masses often result in strengths that
in rock

t
spacin and it
reaction is proportional to the discontinuity spacing, aree less tha of the intact rock. Therefore, the presence
le than that
0°–30°
0° –30 and
decreases gradually within the range of 0°–30° nd tends to d
of these discontinuities creates weakness planes along
increase within the range 30°–90°
–90°.
ge of 30°–90°. which failures may initiate and propagate. The overall

u
beh
behavior of the rock mass is affected by the mechanical
ubgrade reaction modulus  Rock
Keywords Subgrade properties of the intact rock and by the condition of the
foundation  Rock discontinuity
continuity  Model-scale tes
Model-sca footing tests discontinuities.

A
There are several studies for determining the elastic
modulus of the rock mass using empirical correlations
1 Introduction with the rock properties. Heuze (1980) stated that the
modulus of deformation of rock masses ranges between
Soil–foundation interaction is a ch
il–foundation interactio challenging problem in 20 and 60 % of the modulus measured on intact rock
geotechnical engineering. Becau
technical engineeri Because of the complex behavior specimens in the laboratory. Hoek and Brown (1980a, b)
off soil,
oil, the subgrade in so soil–foundation interaction prob- proposed an empirical failure criterion for rock masses
ms is replaced
lems r by a simpler system called a subgrade containing two parameters that are related to the degree of
model. l.. In the practical
prac design of mat foundations, struc- rock mass fracturing. Empirical expressions have also
gineers prefer to model the soil mass as a series of
tural engineers been proposed between those parameters and the rock
i
elastic springs, known as the Winkler foundation. In other quality designation (RQD), the RMR, and the Q ratings.
However, previous studies have focused on the fact that
deformation modulus of rock mass with the discontinuity
& Sangseom Jeong is smaller than that of the intact rock. Because the sub-
soj9081@yonsei.ac.kr grade reaction modulus is proportional to deformation
1 modulus, the subgrade reaction modulus of the rock mass
Department of Civil and Environmental Engineering, Yonsei
University, 50 Yonsei-ro, Seodaemun-gu, Seoul 120-749, with discontinuity is expected to be smaller than that of
Republic of Korea the intact rock.

123
2056 J. Lee, S. Jeong

Table 1 Empirical methods of subgrade reaction modulus (ks)

Proposer Empirical method Application
h i  0:108
Biot (1937) Es B4 Infinite beams resting on an elastic soil continuum
ks ¼ 0:95
B 
Es
1m2s  EI

Terzaghi (1955) Sand Rigid plate placed on a soil medium

h i2
ks ¼ k0:3 ðkN=m3 Þ  B ðmÞþ0:3
2B ðmÞ
ðmÞ
Clay
h i qffiffiffiffiffiffiffiffiffi4ffi
Vesic (1961) 12 Es B Beams resting on elastic half space
B  1v2 
ks ¼ 0:65 Es
EI s

Meyerhof and Baikie (1963) ks ¼ Bð1v

Es
2Þ
Buried circular conduits

y
s

Es ð1vs Þ l
Vlassov (1966) ks ¼ ð1þvs Þð12vs Þ  2B
Beams and plates resting on elastic half space
alf spa

Kloppel and Glock (1979) ks ¼ Bð1þv

2Es
sÞ
Buried circular conduits

p
Selvadurai (1985) ks ¼ 0:65
B  ð1v2 Þ
Es Buried circular conduits
s

o
Since the 1930s, many studies have been performed by q
ks ¼ ðð1Þ
1Þ
many researchers on the subgrade reaction modulus (ks) of d

c
soil (Biot 1937; Terzaghi 1955; Vesic 1961; Meyerhof and where q is thehe soil pressure and d is the
essure at a given point an
Baikie 1963; Vlassov and Leontiev 1966; Kloppel and settlement
nt of the mat at the same point.
Glock 1979; Selvadurai 1985; Horvath 1989; Daloglu and The
he subgrade constant for a
ade reaction modulus is not a cconstan
Vallabhan 2000; Elachachi et al. 2004; Moayed and Naeini

r
factors, such as the
given soil; it depends upon a number of factors
2006). In addition, several empirical methods for soil have width and the shape of the foun foundation in addition to the
been proposed. However, there are very few available ble depth of embedment
mbedment of th the foundation. There are numerous

o
methods of ks for the rock mass compared with thatt of the semi-empirical models tthat can be used to determine the
soil because the available loading test data for or the rockk subgrade reaction modulus aas a function of the elastic
mass were insufficient. modulus ((E),
m E), the Poisson
Poisson’s ratio (m) of the soil, and the

h
The predicted results for the existingg subgrade de reaction footing width ((B).
foot B). Pre
Previous authors have each suggested a
modulus of soil differ from those off rock masses, es, although different
iffere but suitsuitable expression. The empirical methods

t
precise analysis is performed by y empirical methods for soil. are summariz
summarized in Table 1. Many studies (Biot 1937;
Less is known about the subgrade bgrade reaction modulus in Terzaghi 1955; Vesic 1961; Meyerhof and Baikie 1963;
cupies two-thirds of the total land
weathered rock, which occupies Vlass
Vlassov and Leontiev 1966; Kloppel and Glock 1979;

u
area of the Korean n peninsula.
sula. The need for more rresearch S
Selvadurai 1985; Horvath 1989; Daloglu and Vallabhan
on the subgradede reactionon modulus of rock masses has h been 2000) have investigated effective factors and approaches
emphasized.ed. for determining ks.

A
Thisis paper iss intended to evaluate the ssubgrade rereaction Terzaghi (1955) suggested values of ks for a
modulus of jointed weathered rocks. A series of model- 30 cm 9 30 cm rigid slab placed on a soil medium. His
scale
cale footing tests are performed to take in into account var- work showed that the value of ks depends on the dimen-
ious
us factors influencing the su subgrade rea
reaction modulus, i.e., sions of the area acted upon by the subgrade reaction, and
the rock discontinuity spacing and iinclination. Based on the he incorporated size effects in his equations.
obtained
ained results, an appropr
appropriate and simple subgrade For footings on sand,
eaction
ion modulus ((kkj) is proposed,
reaction pr particularly for jointed  
rockk foundations.
foun B ðmÞ þ 0:3 ðmÞ 2
ks ¼ k0:3 ðkN=m3 Þ  ð2aÞ
2B ðmÞ
For footings on clay,
2 Available Methods for Determining  
the Subgrade Reaction Modulus 0:3 ðmÞ
ks ¼ k0:3 ðkN=m3 Þ  ð2bÞ
B ðmÞ
The subgrade reaction modulus (ks) is a mathematical where ks is the desired value of the modulus of subgrade
constant that represents the foundation’s stiffness; it is reaction for full-sized footings, k0.3 is value of k from a
defined as the ratio of the pressure (q) against the mat to the plate load test, B is the footing width.
settlement (d) at a given point,

123
Experimental Study of Estimating the Subgrade Reaction Modulus on Jointed Rock Foundations 2057

Vesic (1961) showed that ks depends upon the stiffness   rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi4ffi

0:65 Es 12 Es  B
of the soil, as well as the stiffness of the structure, so that ks ¼   ; ð3Þ
B 1  v2s EI
similarly sized structures of different stiffnesses will yield
different values of ks for the same applied load. He where Esis the elastic modulus of the soil, ms is Poisson’s
extended Biot’s solution by providing the distributions of ratio of the soil, E is elastic modulus of the beam, I is
deflection, moment, shear and pressure along the beam. He the moment of inertia of the beam.
found that the continuum solution correlated with the Vesic (1961) suggested an equation for ks to use in the
Winkler model Winkler model. For practical purposes, Vesic’s equation
reduces to
UTM (Universal Test Machine)
Es
ks ¼
ð4Þ

y
B 1  v2s
Vlassov and Leontiev (1966) introduced ion ffor
ed an equation

p
beams and plates resting on elasticc half-space, but ut tthe
ambiguities of estimating l in Table
able 1 (a non-dimensional
on-dimension
parameter) make the problem m more complex (Sadrekarim
(Sadrekarimi

o
and Akbarzad 2009). Thee equations
ns given by Meyerhof and
Baikie (1963), Kloppel
ppel and Glock
ck (1979), and Selvadurai
Footing
oposed for computing the horizontal sub-
(1985) were proposed

c
Specimen
Steel ion modulus
grade reaction conduits.
us in buried circular condu
Testing
Box

r
3 Model-Scale
-Scale Footing Tests

In this study, the subgsubgrade reacti

reaction modulus and the

o
UTM Controller bearing capacity of the mat fou foundations considering rock
ddiscontinuities (or joints) we
were investigated by performing
Fig. 1 Test set-up for model-scale footing tests model-scale footing test
mo tests.

t h
Fig. 2 Simple preparation
process of jointed rock
specimens: a preparing
gypsum–sand–water mix,

u
b aluminum mold, c pouring
gypsum–sand–water mix into
ck
the mold and d jointed rock
specimen

A
123
2058 J. Lee, S. Jeong

3.1 Testing Apparatus and Specimen Preparation 3.2 Test Boundary Conditions

Model-scale footing tests were performed in a To study the factors that influence the subgrade reaction
48 cm 9 48 cm 9 28 cm steel box (Fig. 1). The square modulus and the bearing capacity of rock foundations, a
mat foundation, 8 cm wide (B), was made of an aluminum total of 21 model footing tests were conducted on rock
plate 4 cm thick (t). Tests were performed with the foot- specimens under various conditions in consideration of the
ings located at the rock surface. The mat foundation was rock mass discontinuity inclination and spacing. The
loaded vertically at a constant rate of 2 9 10-5 m/s until a effects of the joint roughness, filling material, and dis-
settlement of at least 0.1B occurred. The ultimate bearing continuous friction angle itself were not taken into account.
capacity was defined as the bearing stress that produced a In this study, the unconfined compressive strength (qu), the

y
relative settlement of 0.1B. Although choosing to define discontinuity inclination (Id, where the index d refers
er to the
qult at a relative settlement of s/B is arbitrary, the 0.1B discontinuity or joint), and the discontinuity uity apparent
appa
method is convenient and easy to remember, and it may

p
actually be close to the average soil strain at failure (Cerato
and Lutenegger 2007).
Natural rock blocks with regular discontinuity patterns

o
are required for model-scale footing tests. It is impossible
to perform a number of tests under various boundary
conditions because it is difficult to prepare large rock block

c
samples and to make regular discontinuity patterns with the
rock block. In this study, therefore, two industrial gypsum
plasters were used to make rock-block specimens with

r
regular discontinuity patterns by using a developed
experimental apparatus. These plasters can be molded into to
any shape when mixed with water and sand. The photos tos for

o
the process of preparing artificial jointed rock specimenss
are shown in Fig. 2.
The properties of the artificial rock are similarr to those

h
of typical weathered rocks (Indraratnaatna et al. 1998; Yang

t
and Chiang 2000; Jiang et al.. 2004; Seol et al. al 2008).
psum–sand–water ratio, the
Table 2 summarizes the gypsum–sand–water
unconfined compressive ve strength (q(qu), and the Young’s

u
modulus (Es) of thee rock specimens that were used in this
study.

A
able 2 Material properties of test ssamples
Table
Parameters
ameters Artificial ro
A rock A Artificial rock B

gypsum–sand–water ratio
um–sand–water rat 1.5:1
1.5:1.5:1 2:1:1
CS (MP
UCS (MPa) 15 24
Es (MPa)
MPa)
Pa) 860 1520 Fig. 3 Testing devices and boundary conditions: a plan view and
b front view

Table 3 Summary of test

Variable Values
boundary conditions
Artificial rock A Artificial rock B

UCS (MPa) 15 24
Discontinuity spacing, Sd (cm) Intact, 4, 8, 12, 16 Intact, 4, 8, 12
Discontinuity inclination, Id (°) 0°, 30°, 60°, 90° 90°

123
Experimental Study of Estimating the Subgrade Reaction Modulus on Jointed Rock Foundations 2059

Sd = 16 cm (2.0B) Sd = 12 cm (1.5B) Sd = 8 cm (1.0B) Sd = 4 cm (0.5B)

2cm 2cm 2cm 2cm

4cm
12cm 8cm
16cm 4cm
4cm
28cm 28cm 8cm28cm
4cm 28cm
Id = 0 12cm 4cm
8cm

48cm 48cm 48cm 48cm

16cm 12cm 8cm 8cm 8cm 4cm 4cm 4cm

30º 30º 30º 30º

Id = 30 28cm 28cm 28cm 28cm

y
48cm 48cm 48cm 48cm

16cm 12cm 8cm 8cm 8cm 4cm 4cm 4cm

60º 60º 60º 60º

p
Id = 60 28cm 28cm 28cm 28cm
m

48cm 48cm 48cm 48cm

o
16cm 12cm 8cm 8cm 8cm 4cm 4cm
cm
m 4
4cm
m

Id = 90 28cm 28cm 28cm 28cm

c
48cm 48cm 48cm 448cm
8cm

r
Fig. 4 Rock specimens with joints at various conditions (qu = 15 MPa)

o
spacing (hereafter Sd) were determined to represent the test subgrade reaction (ks) and the ultimate
the modulus of subgrad
boundary conditions. Two unconfined compressive ompressivee (qult) decrease
bearing capacity (q decreased compared with intact rock
strengths (15 and 24 MPa), four discontinuity inuity apparent
p ddue to the rock discontinui
discontinuities.

h
spacings (0.5, 1.0, 1.5, and 2.0 B) and d four discontinuity
scontinuity
inclinations (0°, 30°, 60°, and 90°)) were used as the test 3.3.1 Effect of the Unconfined Compressive Strength (qu)

t
conditions. A summary of the test boundary condi conditions is
given in Table 3. Figure 5a shows the stress-settlement curves from typical
Figure 3 shows thee testing
esting devices and the boundary tests on artificial rock specimens with two unconfined

u
erall dimensions
conditions. The overall boundaries com-
mensions of the boundari compressive strengths (qu = 15 and 24 MPa) under the
co
prise a width of 3.0 times ( from
mes the mat width (B) m the mat same discontinuity spacing [Sd = 12 cm (1.5B)] and
center andd a height equal
qual to 3.5 times the mat width ((B). inclination (Id = 90°). As shown in Fig. 5b, c, with an

A
Thesee dimensions
ons were considered adequate elimin the
adequ to eliminate increase in the unconfined compressive strength (qu), the
nfluence of boundary effects on the m
influence mat perform
performance based modulus of subgrade reaction (ks) increases. In addition,
on (Boussinesq 18
n a review of the literature (Bou 1883). The mat is the ultimate bearing capacity (qult) of the stress–settlement
made
ade of an aluminum plate plate, which aallows for rigorous curves tends to increase as qu increases.
analysis behavior under onl
lysis of the behavi only rock mass conditions,
regardless
rdless of the mat foundatio
foundation properties. 3.3.2 Effect of the Rock Discontinuity Spacing (Sd)

3.3 Test Results and Discussion Figure 6 shows the stress–settlement curves with different
discontinuity spacings [Sd = 4 cm (0.5B), 8 cm (1.0B),
A total off 21 iindividual tests were conducted under various 12 cm (1.5B), 16 cm (2.0B), and intact rock] under two
boundary conditions described in the previous sections. discontinuity inclinations, i.e., 60° and 90°. The stress–
The rock specimens with joints at various conditions are settlement curve tends to be similar to the curve of intact
shown in Fig. 4. In this study, only a selection of typical rock as the discontinuity spacing (Sd) increases. As shown
test results is presented. The modulus of subgrade reaction in Fig. 7, it is noted that the modulus of subgrade reaction
was considered as an initial tangent modulus (around (ks) is proportional to the discontinuity spacing (Sd). In
s = 2 mm) over the estimated working range of bearing addition, qult tends to increase as the discontinuity spacing
pressure. From the results obtained from the model tests, (Sd) increases.

123
2060 J. Lee, S. Jeong

(a) 45 (a) 40
40
35
35
30
Stress (MPa)

Stress (MPa)
30
25
25
20 20

15 qu=24MPa
qu=24MPa Intact
Intact
15
16Sd=16cm (2.0B)
10 qu=15MPa
qu=15MPa
10 12Sd=12cm (1.5B)

y
5
8 Sd=8cm (1.0B)
5
0 4 Sd=4cm (0.5B))
0 1 2 3 4 5 6 7 8
0

p
(b) Settlement (mm) 0 1 2 3 4 5 6 7 8
Modulus of subgrade reaction (ks) (MN/m3)

1600 Settlement (mm)

1400 (b) 40

o
1200
35

c
1000
30

Stress (MPa)
800
25
600 Intact
intact

r
20
Sd=12cm (1.5B)
sd/B=1.5
400 Intact
Intact
sd/B=1.0
Sd=8cm (1.0B) 15 16Sd=16cm (2.0B)
200 sd/B=0.5
Sd=4cm (0.5B)

o
12Sd=12cm (1.5B)
10
0 8 Sd=8cm (1.0B)
12 14 16 18 20 22 24 26 28 5 4 Sd=4cm (0.5B)
Unconfined compressive strength (qu) (MPa)
Pa)

h
(c) 0
0 1 2 3 4 5 6 7 8

t
60
bearing capacity (qult) (MPa)

Settlement (mm)

50 Fig. 6 Stress–settlement curves with different discontinuity spacing:

a discontinuity inclination (Id) = 60°, b discontinuity inclination

u
40 (Id) = 90°

30

A
Intact
intact
shown in Fig. 9, the modulus of subgrade reaction (ks) and
20
Ultimate bea

Sd=12cm (1.5B)
sd/B=1.5 the ultimate bearing capacity (qult) of a rock foundation
100
Sd=8cm (1.0B)
sd/B=1.0 with vertical discontinuity (90°) are greater than that of a
Sd=4cm (0.5B)
sd/B=0.5 rock with inclined discontinuities. For the condition in
0 which the joint inclination is 0°, there is no failure along
12 14 16 18 20 22 24 26 28 discontinuity planes, and failure occurs in rock material. In
Unconfined compressive strength (qu) (MPa) addition, the values of ks and qult decrease gradually within
the range of 0° \ Id \ 30° and tend to increase within the
Effect of unconfined compressive strength (qu): a stress–
Fig. 5 Ef
settlement curves with qu = 15 and 24 MPa [Sd = 12 cm (1.5B), range of 30° \ Id \ 90°. In the case of 30° and 60°, the
Id = 90°], b modulus of subgrade reaction, c ultimate bearing rock mass fails along the plane of rock discontinuities. It is
capacity observed that the least value of ks and qult is in the vicinity
of joint inclination of 30°. When the discontinuity incli-
3.3.3 Effect of the Rock Discontinuity Inclination (Id) nation is 90°, each rock block behaves as an individual
column, and the bearing capacity increases. This trend is in
Figure 8 shows the stress–settlement curves with varying general agreement with previous research regarding the
discontinuity inclinations (Id = 0°, 30°, 60°, and 90°) bearing capacity of the footing (Roy 1993; Sutcliffe et al.
under the same discontinuity spacing (Sd = 12 cm). As 2004; Maghous et al. 2008).

123
Experimental Study of Estimating the Subgrade Reaction Modulus on Jointed Rock Foundations 2061

(a) (a)
1200
Modulus of subgrade reaction (ks) (MN/m3)

Modulus of subgrade reaction (ks) (MN/m3)

1200

1000 1000

800 800

600
600

90Id=90 400 Sd=16cm (2.0B)

S/B=2.0
400

y
0 Id=0 Sd=12cm (1.5B)
S/B=1.5
60Id=60 200 S/B=1.0
Sd=8cm (1.0B)
200
S/B=0.5
Sd=4cm (0.5B)
30Id=30
0

p
0 0 15 30 45 60 75 9
90
0 0.5 1 1.5 2 2.5
Discontinuity Inclination (º)
Discontinuity spacing/mat width ratio (Sd/B)
(b) 40

o
(b)

Ultimate bearing capacity (qult) (MPa)

40
Ultimate bearing capacity (qult) (MPa)

35
35

c
30
30
25
25
20

r
20
15
Sd=16cm (2.0B)
S/B=2.0
15 Id=90
90
10 Sd=12cm (1.5B)
S/B=1.5

o
0Id=0
10 Sd=8cm (1.0B)
S/B=1.0
Id=60
60 5
S/B=0.5
Sd=4cm (0.5B)
5 Id=30
30
0

h
0 15 30 45 60 75 90
0
0 0.5 1 1.5 2 2.5 Discontinuity Inclination (º)

t
Discontinuity spacing/mat
ing/mat width ratio (Sd//B) Fig. 9 Effect of discontinuity inclination (Id): a modulus of subgrade
reaction,
reaction b ultimate bearing capacity
Fig. 7 Effect of discontinuityy spacing (Sd): a modulus
dulus of subgrade

u
a g capac
reaction, b ultimate bearing capacity
apacity
pac ty
4 Proposed Modulus of Subgrade Reaction
for Jointed Rock

A
The modulus of subgrade reaction can be obtained from
40
field and laboratory tests, and semi-empirical equation.
335 Among the semi-empirical methods, Vesic’s model is
widely applied to mat foundations in the literature (Bowles
30
1996). However, these empirical methods were suggested
Stress (MPa)

25 based on the soil; therefore, they do not consider the

influence of rock mass discontinuities.
20
0
A rock foundation is supported by the jointed rock mass,
15 and not by intact rocks. The behavior of foundations on
Id=90
90
0Id=0
rock is largely dependent on the strength of the rock mass.
10
60
Id=60 The rock mass consists of intact rock and discontinuities
5 30
Id=30 (joints or fractures, faults, and possibly bedding planes).
Discontinuities usually have a lower resistance, higher
0
0 1 2 3 4 5 6 7 8 deformability, and conductivity than the intact rock and, in
Settlement (mm) most cases, govern the behavior of the rock mass. Clearly
rock mass discontinuities affect the bearing behavior.
Fig. 8 Stress–settlement curves with different discontinuities Discontinuities may significantly influence the strength of

123
2062 J. Lee, S. Jeong

(a) 40 (c) 40

35 35

30 30
Stress (MPa)

Stress (MPa)
25 25

20 20

15 Intact
Intact 15 Intact
Intact
Id=90
90 Id=90
90
10 0Id=0
10 0Id=0
Id=60
60 Id=60
60
5 5

y
Id=30
30 Id=30
30
0 0
0 1 2 3 4 5 6 7 8 0 1 2 3 4 5 6 7 8
Settlement (mm) Settlement (mm)

p
(b) 40 (d) 40

35 35

o
30 30
Stress (MPa)

Stress (MPa)
25 25

c
20 20

15 Intact
Intact 15 IIntact
Intact
Id=90
90 Id=
90 =90
10 10

r
0Id=0 0Id=0
5 Id=60
60 5 60
Id=60
Id=30
30 30
Id=30
0

o
0 1 2 3 4 5 6 7 8 0 1 2 3 4 5 6 7 8
Settlement (mm) Settlement (mm)

ity inclination: a Sd = 16 cm (2.0B), b Sd = 12 cm (1.5B), c Sd = 8 cm (1.0B) and

Fig. 10 Stress–settlement curves with varying discontinuity

h
d Sd = 4 cm (0.5B)

t
the rock mass, depending on their inclination and the nat- 1
rial in the discontinuities (Pells and
ure of the filling material 0.9

u
Turner 1980; Zhangg and Einstein 1998; Zhang 2010 2010; Jeong
0.8
et al. 2010; Leee et al. 2013).
013). Thus, besides the intact
inta rock
Joint Reduction Factor

0.7
properties, the influencece of rock mass discontinuities
disco mu
must
0.6

A
be takenen into account.
c
The results obtained from the model tests show sho that the 0.5
modulus of subgrade reaction of rock foundafoundation is mainly 0.4
Sd /B=2.0
S/B=2.0
affected
fected by the rock strength and the ro rock mass disconti- 0.3 Sd /B=1.5
S/B=1.5
nuities.
ties. The model-scale ffooting testest results are shown in
0.2 Sd /B=1.0
S/B=1.0
Fig. 10. The modulus of subgrad
subgrade reaction is considered as Sd /B=0.5
S/B=0.5
0.1
ann initial modulus and the results show that a
nitia tangent modulus,
uction
tion in the modulu
reduction modulus of subgrade reaction (ks) occurred 0
0 15 30 45 60 75 90
due to o ro
rock discont
discontinuities. In this study, a joint reduction
Major Discontinuity Inclination (º)
factor (JJf) is proposed considering rock discontinuities
based on the test results. Joint reduction factor (Jf) is Fig. 11 Joint reduction factor (Jf) chart
defined as the ratio of the subgrade reaction modulus of the
jointed rock and that of intact rock. Figure 11 shows a joint
reduction factor (Jf) chart as a function of the discontinuity modulus of subgrade reaction is proportional to the dis-
inclination (Id) and the discontinuity spacing/mat width continuity spacing, and it decreases gradually within the
ratio (Sd/B). The modulus of the subgrade reaction, con- range of 0°–30° and tends to increase within the range of
sidering rock discontinuities, decreased by up to approxi- 30°–90°. Table 4 summarizes the values of the joint
mately 60 %, compared to the intact rock. Additionally, the reduction factor (Jf) based on the obtained results.

123
Experimental Study of Estimating the Subgrade Reaction Modulus on Jointed Rock Foundations 2063

Table 4 Joint reduction factor that the least value of ks is in the vicinity of joint
Sd/B a
Ibd
inclination of 30°.
3. As a result, the joint reduction factor (Jf) is proposed
0 30 60 90 by varying the major factors, i.e., the discontinuity
0.5 0.63 0.43 0.58 0.71 inclination (Id) and the discontinuity spacing/mat width
1.0 0.73 0.58 0.64 0.78 (Sd/B). By taking into account the rock mass discon-
1.5 0.83 0.63 0.71 0.86 tinuity, the proposed subgrade reaction modulus (kj)
2.0 0.85 0.67 0.73 0.91 can be used for jointed weathered rocks.
a
Discontinuity spacing/mat width
Acknowledgments This work was supported by the National

y
b
Major discontinuity inclination
Research Foundation of Korea (NRF) Grant funded by the Korean
government (MSIP) (No. 2011-0030040).
Finally, a subgrade reaction modulus (kj) for jointed

p
rock is proposed. In particular, by considering the discon-
tinuity spacing and inclination, a simple and improved kj is References
introduced in this study. The kj can be expressed by the

o
Biot MA (1937) Bending of infinite
nfinite beams foundation.
ms on an elastic foundation
joint reduction factor (Jf) and subgrade reaction modulus. J Appl Mech Am Socc Mech Eng 59:A1–A7
g 59:A
As a result, kj is suggested as follows: Boussinesq J (1883) Application des potentials a l’etude de
  l’equilibre et due movement
ement des solides elastiques. Gauthier-

c
E
kj ¼ Jf  ks ¼ Jf  ð5Þ aris
Villars, Paris
Bð1  m2 Þ Bowles JE (1996) Foundationation analysis and design, 5th edn. McGr
McGraw-
ill, New York
Hill,
where Jf is the joint reduction factor (Table 4), the other rato AB, Lutenegger
Cerato utenegger AJ (2007) Scale effects of shal
shallow foundation

r
parameters were defined previously. bearing capacity on granular material. J Geot Geote Geoenviron
Geotech
3(10):1192–1202
Eng 133(10):1192–1202
Daloglu AT, Vallabhan CVG (2000) Values oof K for slab on winkler

o
n. J Geotech G
foundation. E 126(5):463-471
Geoenviron Eng
5 Conclusions Elachachi SM, Breysse D, Houy L (2 (2004) Longitudinal variability of
soils and structural response of sewer networks. Comput Geotech
The main objective of this study was too proposee the sub- 31:625–641

h
Heuze FE (1980) Scale effects in the determination of rock mass
Heuz
grade reaction modulus of a rock massss that can consider the str def
strength and deformability. Rock Mech 12:167–192

t
rock discontinuity conditions. Due to the limitations
lim Hoek
ek E, Brown ET (1980) Empirical strength criterion for rock
inherent in field measurements
ntss in Korean rock, a series of masses. ASCE J Geotech Eng 106:1013–1035
experimental studies werere performed to investigate the Hoek E, B Brown ET (1997) Practical estimates of rock mass strength.

u
I J Rock Mech Min Sci 34(8):1165–1186
Int
factors that influence
nce thee subgrade reaction modulus
mod for H
Horvath JS (1989) Subgrade models for soil–structure interaction
ns. The conclusions
rock foundations. onclusions of this study are as analysis. J Found Eng Curr Princ Pract Proc ASCE 20:599–612
follows: Indraratna B, Haque A, Aziz N (1998) Laboratory modeling of shear
behaviour of soft joints under constant normal stiffness condi-

A
1. Based on the results of the model footing tests,te the tions. Geotech Geol Eng 16:17–44
predicting the subgra
influential factors for predic subgrade reaction Jeong SS, Cho HY, Cho JY, Seol HI, Lee DS (2010) Point bearing
modulus ((kks) and the ultima
ultimate bearing capacity (qult) stiffness and strength of socketed drilled shafts in Korean rocks.
Int J Rock Mech Min Sci 47:983–995
for rock foundations are introd introduced. The results Jiang Y, Xiao J, Tanabashi Y, Mizokami T (2004) Development of an
showed that the values of ks and qult are highly automated servo-controlled direct shear apparatus applying a
dependent on ffactors, i.ei.e., the strength of the rock constant normal stiffness condition. Int J Rock Mech Min Sci
mass, the spacing, aand the inclination of the rock
mass 41:275–286
Kloppel K, Glock D (1979) Theoretische und experimentelle
discontinuity.
isco W
With the increase in both the rock untersuchungen zu den traglastproblemen beigewiecher, in die
ength and ddiscontinuity spacing, ks and qult increase.
strength erde eingebetteter rohre. Veröffentlichung des Instituts Statik
2. articu
In particular, the rock mass discontinuity is found to und Stahlbau der Technischen Hochschule Darmstadt, H-10
affect the subgrade reaction modulus of the rock Lee JH, You KH, Jeong SS, Kim JY (2013) Proposed point bearing
load transfer function in jointed rock-socketed drilled shafts. Soil
foundation. Consequently, the discontinuity in the rock Found 53(4):596–606
mass causes the reduction of the modulus of subgrade Maghous S, Bernaud D, Freard J, Garnier D (2008) Elastoplastic
reaction (ks) by up to approximately 60 % compared to behavior of jointed rock masses as homogenized media and finite
the intact rocks. In addition, the value of ks decreases element analysis. Int J Rock Mech Min Sci 45(8):1273–1286
Meyerhof GG, Baikie LD (1963) Strength of steel sheets bearing
gradually within the range of 0°–30° and then tends to against compacted sand backfill. Highway research board
increase within the range of 30°–90°. It is observed proceedings 30

123
2064 J. Lee, S. Jeong

Pells PJN, Turner RM (1980) End-bearing on rock with particular Terzaghi KV (1955) Evaluation of coefficient of subgrade reaction.
reference to sandstone. Proc Int Conf Struct Found Rock, Geotechnique 5(4):297–326
Balkema, Rotterdam 1: 181–190 Vesic AB (1961) Beams on elastic subgrade and Winkler’s hypoth-
Roy N (1993) Engineering behavior of rock masses through study of esis. Proc 5th Int Conf Soil Mech Found Eng 845–850
jointed models. Ph.D thesis, Indian Institute of Technology, Vlassov VZ, Leontiev NN (1966) Beams, plates, and shells on elastic
Delhi, India foundations. Translated from Russian, Israel Program for
Sadrekarimi J, Akbarzad M (2009) Comparative study of methods of Scientific Translations, Jerusalem
determination of coefficient of subgrade reaction. Electron J Yang ZY, Chiang DY (2000) An experimental study on the
Geotech Eng 14:1–14 progressive shear behavior of rock joints with tooth-shaped
Selvadurai APS (1985) Soil–pipeline interaction during ground asperities. Int J Rock Mech Min Sci 37:1247–1250
movement. In: Bennett FL, Machemehl JL (eds) Arctic, civil Zhang L (2010) Method for estimating the deformability of heavily
engineering in the Arctic offshore. ASCE speciality conference, jointed rock masses. J Geotech Geoenviron Eng 136:1242–1250

y
San Francisco, pp 763–773 Zhang L, Einstein HH (1998) End bearing capacity of drilled shafts in
Seol HI, Jeong SS, You KH (2008) Shear load transfer for rock- rock. J Geotech Geoenviron Eng 124:574–584
socketed drilled shafts based on borehole roughness and Ziaie Moayed R, Naeini SA (2006) Evaluation of modulu modulus of
geological strength index (GSI). Int J Rock Mech Min Sci subgrade reaction in gravely soils based on standard penetrati
netra
penetration

p
45(6):848–861 ernational conference
test (SPT). Proceedings of the sixth international ence on
Sutcliffe D, Yu HS, Sloan SW (2004) Lower bound solutions for physical modelling in geotechnics,, 6th ICPMG, MG, Hong KonKong,
bearing capacity of jointed rock. Comput Geotech 31(1):23–36 Chapter 115

c o
o r
u t h
A
123